> > I use a Dell precision
>
> 	What does this means in terms of the graphics card?
> 	Those specifics are probably useful to help replicate.
>
> 	I don't have any dell equipment here (though one of the other
> 	devs might), but depending on the graphics, that might help
> 	zero it in.
>

That means a NVIDIA Quadro FX 880M graphics card

> > I am wondering about the drivers also, but I cannot say for sure. I did upgrade my drivers before posting just to make sure, but still it is strange that I cannot see this behaviour on any other opengl program.
> > I did download and compile freeglut, their demos run smooth without any problems.
>
> 	I see.
>
> 	Well, you could open an STR with all the details we've found to date.
>
> 	It's going to be a problem though if the devs can't replicate.
>
> 	If you're able to identify the issue, supply a patch, or at least let us
> 	know what changes the problem if you try messing around with the FLTK innards.
> 	You may find something in the FLTK opengl win32 initialization code that
> 	needs adjusting.
>

I also suspect that there might be problems replicating this bug, so I will try to do some digging of my own. Since I have built the FLTK library on my own, I might tinker a bit with the source and might run a strace (or windows equiv.) on the glut demos and the fltk demos to see if there might be a difference.

> 	You might try tweaking the freeglut examples to enable features FLTK is
> 	enabling (double buffer mode, etc) to see if you can perhaps replicate
> 	the problem outside FLTK, to see if it's a particular opengl feature causing
> 	the problem.

This will be the first Ill be checking. I'll get back when I have some results.